Complex process simulation calculation system for process industry and construction method

Abstract

The invention relates to the field of process industry calculation, in particular to a complex process simulation calculation system for process industry and a construction method. A process simulation calculation system is divided into an inner iterative calculation level and an outer iterative calculation level through solving the problems in mathematical simulation of the complex structure process industrial process. The internal iterative calculation focuses on numerical solution of an industrial unit model, and the external iterative calculation focuses on modeling and analysis of a flow chart so as to reveal a topological structure of the technological process. In the overall method, internal iteration is continuously called through external iteration, model parameters and a process structure are decoupled, internal iteration calculation is nested in external iteration calculation, and overall solution of an external whole process and all internal models is achieved. The problems of unclear description of complex technological processes, low iteration convergence speed and low algorithm efficiency in external iteration calculation are solved.

Description

Technical Field

[0001] The present invention relates to the field of process industry computing, and in particular to a complex process simulation computing system for process industry and a construction method thereof. Background Art

[0002] Process simulation technology describes the process system of process industry in a mathematical way. By combining with computer technology, it can measure and calculate the material flow, energy flow, etc. in the process system. The process simulation system is composed of a variety of industrial equipment units according to a certain topological relationship. These units can be mathematically described by a nonlinear equation group. The entire process system is a very complex and large nonlinear equation group obtained by combining the mathematical models of each unit according to a certain relationship. The solution process of this equation group is the realization process of mathematical simulation of the process system.

[0003] The mathematical model is the core component of the process simulation system, which is mainly composed of two parts: the industrial unit model and the structural model. The industrial unit model describes the specific industrial unit within the system, while the structural model describes the complex network relationship between the various units in the system. The process simulation platform is composed of structures such as the basic physical property database, the thermodynamic method library, the unit mechanism model library, and the calculation system. Among them, the calculation system refers to the mathematical model of the corresponding unit after obtaining the flow chart in the process simulation platform drawing interface, calling the physical property data and thermodynamic methods related to each module from the underlying database, and identifying the process topology structure, determining the calculation order of each module and the value of each logistics, and then calling the appropriate calculation method from the calculation method library to calculate and solve the overall mathematical model of the process. A set of mathematical models of a mature process often has the characteristics of large calculation amount and difficult solution, and the equipment, materials, and structure of the process are often complex and changeable. In the system process design and optimization, it is often necessary to adjust the process. In addition, replacing equipment, adjusting logistics data or changing process structure based on equipment selection makes it more difficult to calculate and solve the overall mathematical model, and the accuracy of the calculation and solution results will directly affect the reliability of the overall process simulation.

[0004] The process simulation calculation system of the prior art mostly adopts the sequential module method (such as Aspen Plus, PRO / Ⅱ, etc.), compiles the mathematical model of each process unit into a solution program, and then performs simulation calculations from front to back in the order of connection of the modules in the process system. When encountering a cyclic material flow or energy flow, it is iterated and converged. However, the process simulation platforms such as UniSim and gPROMS adopt a technical system based on the simultaneous equation method. The basic idea of ​​the simultaneous equation method is to combine all equations describing the process system (including the model equations inside the unit) together, which is mathematically regarded as a very large nonlinear equation group, whose variables include all internal and external variables, namely physical property variables, unit model variables (internal variables), stream variables describing the system structure, and process parameters (external variables). Once such a nonlinear equation group containing all internal and external variables is established, the solution of the steady-state simulation can be regarded as a pure mathematical problem that is in principle unrelated to the process system structure, and even unrelated to the chemical engineering background. The calculation system of the sequential module method has the advantages of being in line with the intuition of engineers, easy to realize generalization, easy to diagnose errors and debug, but its nested iteration calculation efficiency is low. The simultaneous equation method is divorced from the process system structure and process background, and simply simplifies the problem into a mathematical solution problem. Therefore, this method does not conform to the production environment and the intuition of engineers. When the process fails or the simulation fails, it is difficult to diagnose the error. When performing iterative calculations, it is more difficult to estimate the initial value of the variable because it is divorced from the actual production, which increases the difficulty of calculation convergence. Moreover, when the process changes, the simultaneous equation method is difficult to inherit the existing modules, and it is impossible to simplify the calculation process and improve the calculation efficiency by migrating the original method.

[0005] Chen Huidong proposed a computational framework for a process simulation platform based on the sequential modular method and graph theory. By developing a configuration tool, he realized the conversion of a directed graph into an adjacency matrix, and used graph theory methods to perform topological decomposition and processing on the process system, realizing the chemical system segmentation, cutting and sorting methods, and completing the solution of the entire process of the chemical system. After calculating and sorting the process flow and cutting off the reflux stream, the computational convergence method of the cut stream was a hybrid calculation of the direct iteration method and the Wegstein method. The idea of ​​the Wegstein method is similar to the secant method. When solving multivariable nonlinear equations, it often has the characteristics of oscillation convergence. It can only be applied to weak cross-linking, and its application range is relatively narrow. However, this method has a slow iterative convergence speed and low algorithm efficiency.

[0006] From the above, we can see that structural complexity is an important factor in the complexity of mathematical models of process industries. In system process design or optimization, it is often necessary to adjust the process, that is, to change the topology of the system. Because of some small structural changes, the entire system is re-divided, which will greatly reduce the calculation efficiency. Therefore, how to build an accurate, feasible and universal computing system and improve the efficiency of the algorithm is the top priority of process simulation technology, which is of great significance for process development, design and optimization. Summary of the invention

[0007] In view of the above problems, the purpose of this application is to develop a general and effective method for simulating complex structure process processes based on the relevant theories of mathematical optimization and system topological decomposition for several problems faced in the mathematical simulation of industrial processes of complex structure processes. This method divides the process simulation calculation system into two levels: internal iterative calculation and external iterative calculation. The internal iterative calculation focuses on the numerical solution of industrial unit models, while the external iterative calculation focuses on the modeling and analysis of flow charts to reveal the topological structure of the process flow. In the overall method, the internal iteration is continuously called by the external iteration, the model parameters and process structure are decoupled, and the internal iterative calculation is nested in the external iterative calculation to achieve the overall solution of the external full process and each internal model, solving the problems of unclear description of complex process flows, slow iterative convergence, and low algorithm efficiency in external iterative calculations. Through this strategy, we can accurately solve the mathematical models of each unit, effectively solve the problem of the order of calculation of the whole process in the complex process simulation process, and determine the value of logistics between upstream and downstream equipment. Therefore, this method provides a computing system suitable for complex process flow structures, and provides computing technology and algorithm foundation for the development of interactive simulation technology for process industries.

[0008] Based on the above purpose, the present application provides a method for constructing a complex process simulation calculation system for process industry, including two calculation levels: internal iteration calculation and external iteration calculation. The internal iteration is used to calculate the basic model, and the external iteration realizes the whole process flow solution by calling the internal iteration model parameters. Specifically, the following steps are included:

[0009] (1) Identify the target topology diagram and generate an adjacency matrix: Based on the target topology diagram, the user automatically numbers the units and directed lines in the order of module creation when drawing, obtains the initial unit and directed line calculation order, and sets connection nodes at the midpoints of the edges of each unit and the endpoints of the directed lines, associates the basic modules in the internal iterative calculation, and establishes a two-dimensional structure data table of n*3 different units and directed lines. Then, the unit and directed line structure data tables are identified respectively to generate an adjacency matrix;

[0010] (2) Loop identification and sorting algorithm: The adjacency matrix is ​​input in the order of the initial unit calculation sorting, and the adjacency matrix is ​​processed row by row through loop identification. Then, the updated adjacency matrix after the forward processing is reversely processed to obtain the processed unit calculation sorting and the processed adjacency matrix. According to the processed unit calculation sorting, the directed line structure data table is traversed to obtain the reverse order flow stock set of the directed line. Then, according to the processed adjacency matrix, the upstream and downstream reflux identification method is used to determine the final calculation sorting and the flow stock set that needs to be cut off;

[0011] (3) Solving the entire process: Create virtual streams to replace the streams that need to be cut off, then enter the initial variable data and the initial values ​​of the virtual stream variables in the unit structure data table to obtain the initialized stream data table, and call the basic models of each unit calculated by other variable values ​​or internal iterations in sequence to perform full-process calculation and solution; after the initial solution, use the quasi-Newton method to iterate and converge the cut streams. After completing the iterative calculation of all cut streams, output the final stream variable data table to complete the entire process solution.

[0012] Furthermore, the loop identification of step (2) processes forward row by row and updates the adjacency matrix specifically as follows: the cell corresponding to the row of the nearest non-zero element starting from the i-th element of the i-th column upward is denoted as K, and the cell corresponding to the column of the farthest non-zero element starting from the i-th element of the i-th row to the left is denoted as P; according to the initial cell calculation sorting order, it is determined row by row whether there is a backflow in the adjacency matrix. If the adjacency matrix has a downstream cell sorted before cell I, that is, the sum n of all elements from the i-th element of the i-th row to the left is not 0, then there is a backflow; at this time, the sum of all elements from the i-th element of the i-th column upward is m, and it is determined whether m is less than n. If so, the cell I corresponding to the cell calculation sorting and the adjacency matrix is ​​inserted before cell P; if not, it is inserted after cell K, and all adjacency matrices are traversed to obtain the adjacency matrix after forward processing.

[0013] Furthermore, the method for constructing the reverse order flow stock set of directed lines in step (2) is: traverse the directed line structure data table, if the direction from the starting unit to the final unit of a directed line is in reverse order in the unit number sorting, then it is a directed line reverse order flow stock, and all directed line reverse order flow stocks form a directed line reverse order flow stock set.

[0014] Furthermore, in step (2), the upstream and downstream backflow identification method is used to confirm the final calculation order and the set of flow streams that need to be cut off. Specifically, each directed line reverse flow stream is found in the final adjacency matrix, and processed one by one. Assuming there are n directed lines, a set of cuttable flow streams is created with a length of n, and the first one is E→D. Then, the upstream and downstream backflow identification method is used to find the circular flow including the reverse flow stream. If there is only one L1 circular flow including the reverse flow stream, all the flow streams in the circular flow are added to the cuttable flow stream set 1 (there are n sets in total). If there are multiple L1, L2, etc. circular flows including the reverse flow stream, it is determined whether the common flow streams of these circular flows are only E→D. If not, these common flow streams are added to the cuttable flow stream set 1, and then the judgment of the next flow stream is entered. After all the flow streams are judged, the cuttable flow stream set is output.

[0015] Furthermore, the whole process calculation and solution described in step (3) is specifically as follows: when calculating each unit in sequence, the mathematical model of the corresponding unit is called in turn, the flow stream flowing into the unit is determined through the structure data table, and the variable values ​​required to solve the model are obtained from the flow stream variable data table. After the solution is completed, the variable values ​​obtained by the solution are filled into the flow stream variable data table, and the flow stream variable data table is updated and used for subsequent calculations.

[0016] Furthermore, the iterative convergence of the cut-off stream using the quasi-Newton method described in step (3) is specifically as follows: after completing the initial full-process calculation and solution, the values ​​of the two virtual streams obtained at the stream cut-off are set to be equal one by one, and a set of nonlinear equations F(X)=X is obtained, which is transformed into G(X)=F(X)-X=0, and the initial value of the virtual stream is obtained from the stream variable data table, and brought into the quasi-Newton solution, and finally the return stream value X is obtained. The cut-off stream is restored in the stream variable data table, the corresponding virtual stream is deleted, and the return stream value update table is entered.

[0017] Furthermore, after iterative calculation of all cut-off streams in step (3), when there is an iterative loop inside the model and the calculation requires multi-level iterative interaction inside and outside, the full-flow program solution method is integrated into the iterative solution method at the broken stream, and the residual is used as the objective function to achieve iterative convergence processes at different levels.

[0018] And, a complex process simulation computing system for process industry, comprising:

[0019] Internal iteration module: This module relies on the amount of physical property data, including the thermodynamic data and transfer property data of the material, and is used for the calculation and solution of the basic model.

[0020] External iteration module: used to perform mathematical modeling and in-depth analysis on the input flowchart to implement the above steps, wherein it can call the calculation and solution results of the internal iteration module as needed.

[0021] A computer device, comprising:

[0022] Memory; and

[0023] The processor is configured to execute the above-mentioned complex process simulation calculation construction method for process industry based on the instructions stored in the memory.

[0024] A computer-readable storage medium stores computer instructions, which, when executed by a processor, can implement the above-mentioned complex process simulation calculation construction method for process industry.

[0025] Beneficial effects of the present application: The present invention divides the process simulation calculation system into two levels: internal iterative calculation and external iterative calculation. The internal iterative calculation is mainly responsible for the calculation and solution of basic physical properties, thermodynamic methods and unit mechanism models. It integrates more than 20 process industry unit mechanism models for calculating material flow and energy flow, which is sufficient to cover the unit models commonly used in process simulation. The external iterative calculation is based on topological recognition theory and adjacency matrix model, and mainly performs mathematical modeling and in-depth analysis on the input flow chart. It can quickly respond to complex variable topological structures and make accurate process cutting and calculation sorting judgments. This method overcomes the problems of unclear description of complex process flows, slow iterative convergence speed and low algorithm efficiency under the existing process simulation calculation framework by decoupling the model parameters and process structure of the process simulation calculation system, and achieves the calculation and solution effect of the current mainstream process simulation software. This method provides a powerful automated calculation tool for engineers and researchers engaged in process industry simulation, and provides a computing technology and algorithm foundation for the development of interactive simulation technology for process industries. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 It is a structural design diagram of a complex process simulation computing system of the present invention;

[0027] Figure 2 It is a structure diagram of a directed graph recognition and matrix reading algorithm of an external iterative computing system of the present invention;

[0028] Figure 3 It is a forward step diagram of the loop identification and sorting algorithm based on the adjacency matrix of the present invention;

[0029] Figure 4 It is a reverse step diagram of the loop identification and sorting algorithm based on the adjacency matrix of the present invention;

[0030] Figure 5It is a flow stream cutting process step diagram based on the loop identification and sorting algorithm of the adjacency matrix of the present invention;

[0031] Figure 6 This is a technical schematic diagram of the first solution method for solving the whole process stream of the present invention;

[0032] Figure 7 This is a technical schematic diagram of the second solution method for the whole process stream of the present invention;

[0033] Figure 8 This invention uses a complex Sargent and Westberg directed graph as an input case of the present computing system;

[0034] Fig. 9 This is a schematic diagram of the overall case of Experimental Example 2. DETAILED DESCRIPTION

[0035] The present invention is further described in detail below in conjunction with the accompanying drawings. The following specific implementation steps can enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any form.

[0036] First of all, it should be noted that the use of computer simulation to generate a three-dimensional rock mass fracture network model and three-dimensional block identification belongs to the prior art. The technical solution of the present invention is based on the block data obtained by three-dimensional block identification to process small blocks, which will involve a large number of prior arts, and its definition or concept is already well known or mastered in the art. Therefore, unless there is a special explanation in the present invention, other meanings that are consistent with the existing known meanings will not be described one by one in the present invention.

[0037] Example 1 A method for constructing a complex process simulation computing system for process industry

[0038] like Figure 1 As shown, the system includes internal iterative calculation and external iterative calculation, wherein the internal iterative calculation is used to calculate the basic model, including the conventional basic physical property database in the chemical industry, the thermodynamic method library and the unit mechanism model library, etc.; the external iteration realizes the whole process flow solution by calling the internal iterative model parameters, specifically including the identification of the topological structure diagram and the generation of the adjacency matrix, the loop identification and calculation sorting algorithm, and the whole process flow solution.

[0039] 1. Identify the target topology diagram and generate an adjacency matrix: Based on the target topology diagram, the user automatically numbers the units and directed lines in the order of module creation when drawing, obtains the initial unit and directed line calculation order, and sets connection nodes at the midpoints of the edges of each unit and the endpoints of the directed lines, associates the basic modules in the iterative calculation, and establishes a two-dimensional structure data table of n*3 different units and directed lines. Then, the unit and directed line structure data tables are identified respectively to generate an adjacency matrix.

[0040] The number is automatically assigned by the program when the interface is drawn.

[0041] Among them, the target topology structure diagram is a diagram formed in the actual operation of the process industry. It is a module created by the user in the drawing interface according to the target flow chart, such as the Sargent and Westberg directed graph.

[0042] It should be noted that the unit structure table and the directed line structure table are generated adjacency matrices (such as Figure 1 The output matrix N and the output matrix M shown are the same matrix, and both are generated at the same time to compare and verify the correctness of the data; that is, Figure 1 , N=M.

[0043] Its construction process is as follows Figure 2 As shown, it includes two parts: identifying the directed graph to obtain the structure data table and identifying the structure data table to obtain the adjacency matrix, specifically:

[0044] Step (1) construct the unit and directed line numbers and create connecting nodes:

[0045] When creating unit modules, they are numbered with capital letters A to Z in the order of creation. If they exceed the range, they are numbered with AA to AZ, and so on without an upper limit. When creating directed line modules, they are numbered starting with 1 in the order of creation, and there is no upper limit. When the mouse hovers over any module, its corresponding number will be displayed immediately. In addition, each unit's edge midpoint and the endpoint of the directed line are equipped with connection nodes, and the modules associated with these nodes are their parent classes.

[0046] Step (2) Obtain the unit and directed line structure data table:

[0047] 1) Unit structure data table: Create a two-dimensional data structure table with different numbers of units n*3, with the header fields being the unit number "name", the directed line "linein" for the input unit, and the directed line "lineout" for the output unit. First, fill the unit numbers into the "name" column in sequence, traverse the connection nodes, find the directed line number connected to the unit at the arrow end and fill it into "linein", find the directed line number connected to the unit at the starting end and fill it into "lineout", and then output the table after completion.

[0048] 2) Directed line structure data table: Create a two-dimensional data structure table with different numbers of units n*3, and the header fields are the directed line number "name", the starting unit "start", and the end unit "end". First, fill the directed line number into the "name" column in sequence, traverse the connection nodes, find the unit number connected to the starting end of the directed line and fill it into the "start" column, find the unit number connected to the arrow end of the directed line and fill it into the "end" column, and output the table after completion.

[0049] Step (3) Identify the structure data table to generate an adjacency matrix:

[0050] 1) Identify the unit structure data table: first determine the number and name of the elements in name. If the number is n, create an n*n matrix with all zeros. Then extract the numbers in each element in lineout one by one and record the corresponding name element number x (x indicates which name element the element is). For each number, traverse linein, find the same number, and record the corresponding name element number y (which has the same meaning as x). Then fill the number 1 into the (x, y) position of the established matrix until all elements in lineout are traversed, end the loop, and finally output the matrix.

[0051] 2) Identify the directed line structure data table: first determine the number m of different elements contained in start and end, and number them in sequence, establish an m*m matrix, fill it with the number 0, and then extract the corresponding elements in start and end row by row, find the number x corresponding to the start element and the number y corresponding to the end element in the same row, and modify the (x, y) position of the matrix to the number 1. If one of the start and end elements in the row is empty, skip the row, end the loop after traversing all rows, and output the matrix.

[0052] 2. Loop identification and sorting algorithm: Input the adjacency matrix in the order of the initial unit calculation sorting, forward process and update the adjacency matrix row by row through loop identification, and then reversely process the updated adjacency matrix after forward processing to obtain the final unit calculation sorting and the final adjacency matrix. According to the final unit calculation sorting, traverse the directed line structure data table to obtain the reverse order flow stock set of the directed line, and then use the upstream and downstream reflux identification method according to the final adjacency matrix to determine the final calculation sorting and the flow stock set that needs to be cut off.

[0053] like Figure 3 The figure shows the forward processing and updating of the adjacency matrix row by row:

[0054] Step (1) is to number the adjacency matrix:

[0055] Add labels to each row from top to bottom and each column from left to right of the adjacency matrix according to the numbers sorted by the cell numbers. The labels are the cell names corresponding to each row or column.

[0056] Step (2) Identify the adjacency matrix and determine whether there is a backflow by identifying whether there is a downstream unit before sorting: if the critical matrix has a downstream unit sorted before unit I, that is, determine row by row in label order whether the sum of all elements n (i.e., the number of downstream units) from the i-th element to the left of the i-th element in the i-th row is 0. If it is 0, skip the row. If not 0 (there is a backflow), go to step (3), and after traversing all rows, go to step (6);

[0057] Step (3) compares the number of upstream units and downstream units before each unit is sorted:

[0058] The sum of all the elements from the ith element in the ith column of the adjacency matrix upward is recorded as m (the number of upstream units), and n is compared with m. If n is less than or equal to m, go to step (4). If n is greater than m, go to step (5);

[0059] Step (4) handles the case where the number of downstream units is not greater than the number of upstream units:

[0060] The labels added in step (1) record the cell K corresponding to the row with the nearest non-zero element starting from the i-th element in the i-th column of the adjacency matrix, and the cell I corresponding to the i-th row, and insert the cell I after K in the 'Item' in the cell number sorting, and the cells behind it are automatically postponed, and the 'ID' in the cell number sorting remains unchanged, and the previous cell number sorting is replaced by the reordered cell number sorting. The labels initialized for each row and column of the adjacency matrix do not change with their position, and they are transformed in two steps: first, the row corresponding to cell I is inserted below the row corresponding to cell K, and the rows below them are automatically moved to the next row; second, the column corresponding to cell I is inserted to the right of the column corresponding to cell K, and the columns to the right of the columns corresponding to cell I are deleted and automatically moved to the right by one column. After completing the transformation of the adjacency matrix, replace the current adjacency matrix with the transformed adjacency matrix, return to step (2) and continue judging from the i+1th row;

[0061] Step (5) handles the situation where the number of downstream units is greater than the number of upstream units:

[0062] The label added in step (1) records the cell P corresponding to the column corresponding to the first non-zero element from left to right in the i-th row of the adjacency matrix (the column with the farthest non-zero element from the i-th element of the i-th row to the left), and the cell I corresponding to the i-th row, and inserts the cell I before the cell P in the 'Item' in the cell number sorting, and the cell P and the cells behind it are automatically postponed, and the 'ID' remains unchanged. The labels initialized for each row and column of the adjacency matrix do not change with their position, and they are transformed in two steps: first, the row corresponding to the cell I is inserted above the row corresponding to the cell P, and the rows below them are automatically moved to the next row; then the column corresponding to the cell I is inserted to the left of the column corresponding to the cell P, and the columns to the right of them are automatically moved to the right by one column. After completing the transformation of the adjacency matrix, the transformed adjacency matrix is ​​output and replaces the current adjacency matrix, and the step (2) is returned to continue judging from the i+1th row; finally, the adjacency matrix after forward processing is formed.

[0063] Among them, the inverse processing process of the updated adjacency matrix after forward processing is as follows Figure 4 As shown, it is the same as the forward processing step (1) to step (5), except that the processing order is reversed. After the reverse processing, the final unit calculation order and the final adjacency matrix are obtained. According to the final unit calculation order, the directed line structure data table is traversed to obtain the reverse order flow stock set of the directed line. Then, according to the final adjacency matrix, the upstream and downstream reflux identification method is used to determine the final calculation order and the flow stock set that needs to be cut off, as shown in Figure 5 As shown, specifically:

[0064] Step (1) obtains the final unit calculation order and identifies the reverse order stream:

[0065] Traverse all directed lines in the directed line structure data table. If the direction from the starting unit to the end unit of a directed line is in reverse order in the unit number sorting, add the directed line to the reverse flow stock set.

[0066] Step (2) finds each reverse order flow in the final adjacency matrix and processes them one by one. Assuming there are n reverse order flows, create a set of cuttable flow with a length of n, the first of which is E→D. Then, find the circulating flow including the reverse order flow by the upstream and downstream return flow identification method. If there is only one L1 in the circulating flow including the reverse order flow, then add all the flows in the circulating flow to the cuttable flow set 1 (there are n sets in total). If there are multiple L1, L2, etc. in the circulating flow including the reverse order flow, then determine whether the shared flow of these circulating flows is only E→D. If not, add these shared flows to the cuttable flow set 1, and then proceed to the judgment of the next flow. After judging all the flows, output the cuttable flow set.

[0067] Step (3) pops up a dialog box, in which the user can select a specific stream to be cut from the provided set of streams that can be cut. The user then returns the result to the program to determine whether to change the stream to be cut. If the stream is changed, the corresponding calculation order is changed, and the final calculation order and the set of streams that need to be cut are output.

[0068] 3. Solve the whole process flow stream: Create a virtual flow stream to replace the flow stream set that needs to be cut off for cutting operation, then enter the initial variable data and the initial value of the virtual flow stream variable in the unit structure data table to obtain the initialized flow stream data table, and call other variable values ​​or each unit basic model calculated by internal iteration in sequence to perform full process calculation and solution; after the initial solution, use the quasi-Newton method to iterate and converge the cut flow streams, and after completing the iterative calculation of all cut flow streams, output the final flow stream variable data table to complete the whole process solution.

[0069] like Figure 6 As shown, the flow stream set to be cut off, the structure data table, the flow stream variable data, the initial value of the cut flow stream variable, the environment variable, the constraint variable, the mathematical model of each unit (external call), and the whole process calculation sequence are used as input, and the output is the value of each flow stream variable. The method steps are as follows:

[0070] Step (1) obtains the set of streams to be cut obtained by the above method and performs a stream cutting operation:

[0071] Traverse the unit structure data table and the directed line structure data table, delete the streams included in the stream set that needs to be cut off, and replace them with corresponding virtual streams. For example, if you find stream 1 in the stream set that needs to be cut off in the unit structure data table, delete it and replace stream 1 in lineout with virtual stream 1a, and replace stream 1 in linein with virtual stream 1b; if you find stream 1 in the stream set that needs to be cut off in the directed line structure data table, delete the corresponding row, and add the rows corresponding to virtual stream 1a and virtual stream 1b, where the start unit of 1a is the original start unit of stream 1, and the end unit is empty, and the start unit of 1b is empty, and the end unit is the original end unit of stream 1.

[0072] Step (2) Obtain the stream variable data and initialize the stream variable data table including the virtual stream:

[0073] The table has n rows and m columns, where n is the number of streams, each row contains all variable information in its corresponding stream, and m is the number of variables in the stream with the most variables. The user needs to fill in the known variable data, and set the initial value of the virtual stream generated after cutting according to the manual or experience, and add it to the corresponding position in the stream variable data table.

[0074] Step (3) obtains other variable values, such as the environmental variable values ​​required for calculation and the constraint variables that meet the design requirements.

[0075] Step (4) obtains the whole process calculation sequence and the mathematical model of each unit obtained by processing, and performs the whole process calculation and solution:

[0076] When calculating each unit in sequence, the mathematical model of the corresponding unit is called in turn, the flow stream flowing into the unit is determined through the structure data table, and the variable values ​​required to solve the model are obtained from the flow stream variable data table. After the solution is completed, the variable values ​​obtained are filled into the flow stream variable data table, and the flow stream variable data table is updated and used for subsequent calculations.

[0077] After the initial full-process calculation and solution is completed in step (5), the quasi-Newton method is used to iterate and converge the cut-off stream:

[0078] After the whole process is solved, the values ​​of the two virtual streams obtained at the stream cut-off point are set to be equal one by one, and a set of nonlinear equations F(X)=X is obtained, which is transformed into G(X)=F(X)-X=0, and the initial value of the virtual stream is obtained from the stream variable data table, and brought into the quasi-Newton solution, and finally the return stream value X is obtained. The cut stream is restored in the stream variable data table, the corresponding virtual stream is deleted, and the return stream value is entered into the update table.

[0079] It should be noted that when there is an iterative loop inside the model and the calculation requires multi-level iterative interaction between the inside and outside, the full flow program solution method is integrated into the iterative solution method at the fracture stream, and the residual is used as the objective function, such as Figure 7 As shown, the iterative convergence process at different levels is achieved.

[0080] Test Example 1

[0081] The complex process simulation calculation system constructed by the present invention uses data in a fixed format as the basic input of the internal iterative calculation system, and uses unit and directed line structure data tables as the basic input of the external iterative calculation system. It can quickly construct process simulation networks of different complexities by dragging and dropping, and return correct output results. Figure 8 The present invention uses a complex Sargent and Westberg directed graph as an input case of the computing system and obtains a good verification effect.

[0082] Test Example 2

[0083] The complex process simulation calculation system constructed by the present invention is applied to the three-stage flash process in the chemical industry. This calculation system is based on externally constructed flash tanks, mixer mathematical models and related physical properties and thermodynamic databases, and takes the structured process system data shown in Table 1 as input. A three-stage flash directed graph is constructed in the drawing interface by dragging and dropping. The structural data table is obtained by identifying the directed graph topological structure, and then the corresponding adjacency matrix is ​​obtained. The final calculation sorting and cutting of the flow streams are obtained by identifying the adjacency matrix. According to the calculation sorting and cutting of the flow streams, sequential solutions and iterative calculations are performed according to the full-process calculation framework provided by the calculation system, and finally the results of various variables of each flow stream in the whole process and the specific gasification rate of each flash tank are obtained. The result data are shown in Tables 2 and 3, which verify the effectiveness of the calculation system. The overall case diagram is shown in Fig. 9 As shown, this case proves that the complex process simulation calculation system constructed by the present invention has good application effect when it comes to actual process system calculations, and is a feasible technical solution to solve complex process simulation calculations.

[0084] Table 1 Structured process data

[0085]

[0086] Table 2 Specific gasification rate solution results of each flash tank

[0087] Solution Flash Evaporator B2 Flash Evaporator B3 Flash Evaporator B4 Gasification rate 0.522 0.617 0.331

[0088] Table 3 Results of variables for each flow in the whole process

[0089] Stream\Variable Molar flow rate (mol / h) Butane mole fraction Pentane mole fraction Hexane mole fraction 1 453.6 0.3 0.4 0.3 2 706.451 0.283 0.426 0.291 3 368.502 0.398 0.426 0.176 4 337.949 0.157 0.425 0.417 5 141.085 0.231 0.474 0.295 6 111.766 0.279 0.470 0.250 7 227.417 0.502 0.397 0.101 8 226.183 0.097 0.403 0.500 9 252.851 0.252 0.472 0.275

[0090] Note: The actual scope of the present invention includes not only the specific embodiments disclosed above, but also all equivalent solutions that implement or perform the present invention under the claims.

Claims

1. A method for constructing a complex process simulation computing system for process industry, characterized in that: It includes two calculation levels: inner iteration calculation and outer iteration calculation. The inner iteration is used to calculate the basic model, and the outer iteration realizes the whole process flow solution by calling the inner iteration model parameters. It specifically includes the following steps: (1) Identify the target topology graph and generate an adjacency matrix: automatically number the units and directed lines of the target topology graph according to the order in which the modules are created, obtain the initial unit and directed line calculation order, and set connection nodes containing connection information at the midpoint of each unit's edge and the end point of the directed line. Based on this, establish a two-dimensional structure data table of n*3 different units and directed lines, and then identify the unit and directed line structure data tables to generate an adjacency matrix; (2) Loop identification and sorting algorithm: Input the adjacency matrix in the order of the initial unit calculation sorting, and process and update the adjacency matrix row by row through loop identification. Then, reversely process the updated adjacency matrix after forward processing to obtain the processed unit calculation sorting and the processed adjacency matrix. According to the processed unit calculation sorting, traverse the directed line structure data table to obtain the reverse order flow stock set of the directed line. Then, according to the processed adjacency matrix, use the upstream and downstream reflux identification method to determine the final calculation sorting and the flow stock set that needs to be cut off. (3) Solving the entire process: Create a virtual stream to replace the stream set that needs to be cut off, then enter the initial variable data and the initial value of the virtual stream variable in the structure data table to obtain the initialized stream data table, and call the other variable values ​​or each unit basic model calculated by internal iteration in sequence to perform full process calculation and solution; after the initial solution, use the quasi-Newton method to iterate and converge the cut streams. After completing the iterative calculation of all cut streams, output the final stream variable data table to complete the entire process solution.

2. The construction method according to claim 1, characterized in that: The loop identification of step (2) processes forward row by row and updates the adjacency matrix specifically as follows: the cell corresponding to the row of the nearest non-zero element starting from the i-th element of the i-th column upward is denoted as K, and the cell corresponding to the column of the farthest non-zero element starting from the i-th element of the i-th row to the left is denoted as P; according to the initial cell calculation sorting order, it is determined row by row whether there is a backflow in the adjacency matrix. If the adjacency matrix has a downstream cell sorted before the I cell, that is, the sum n of all elements from the i-th element of the i-th row to the left is not 0, then there is a backflow; at this time, the sum of all elements from the i-th element of the i-th column upward is m, and it is determined whether m is less than n. If so, the cell I corresponding to the cell calculation sorting and the adjacency matrix is ​​inserted before the cell P; if not, it is inserted after the cell K, and all adjacency matrices are traversed to obtain the adjacency matrix after forward processing.

3. The construction method according to claim 1, characterized in that: The method for constructing the reverse flow stock set of directed lines in step (2) is: traverse the directed line structure data table, if the direction from the starting unit to the final unit of a directed line is in reverse order in the unit number sorting, then it is a directed line reverse flow stock, and all directed line reverse flow stocks form a directed line reverse flow stock set.

4. The construction method according to claim 1, characterized in that: In step (2), the upstream and downstream backflow identification method is used to confirm the final calculation order and the set of flow streams that need to be cut off. Specifically, each directed line reverse flow stream is found in the final adjacency matrix and processed one by one. Assuming there are n directed lines, a set of cuttable flow streams is created with a length of n, and the first one is E→D. Then, the upstream and downstream backflow identification method is used to find the circular flow including the reverse flow stream. If there is only one L1 circular flow including the reverse flow stream, all the flow streams in the circular flow are added to the cuttable flow stream set 1 (there are n sets in total). If there are multiple L1, L2, etc. circular flows including the reverse flow stream, it is determined whether the common flow streams of these circular flows are only E→D. If not, these common flow streams are added to the cuttable flow stream set 1, and then the judgment of the next flow stream is entered. After all the flow streams are judged, the cuttable flow stream set is output.

5. The construction method according to claim 1, characterized in that: The whole process calculation and solution described in step (3) is specifically as follows: when calculating each unit in sequence, the mathematical model of the corresponding unit is called in turn, the flow stream flowing into the unit is determined through the structure data table, and the variable values ​​required to solve the model are obtained from the flow stream variable data table. After the solution is completed, the variable values ​​obtained by the solution are filled into the flow stream variable data table, and the flow stream variable data table is updated and used for subsequent calculations.

6. The construction method according to claim 1, characterized in that: The specific method of iteratively converging the cut-off stream using the quasi-Newton method in step (3) is as follows: after completing the initial full-process calculation and solution, the values ​​of the two virtual streams obtained at the stream cut-off are set to be equal one by one, and a set of nonlinear equations F(X)=X is obtained, which is transformed into G(X)=F(X)-X=0, and the initial value of the virtual stream is obtained from the stream variable data table, and brought into the quasi-Newton solution, and finally the return stream value X is obtained. The cut-off stream is restored in the stream variable data table, the corresponding virtual stream is deleted, and the return stream value update table is entered.

7. The construction method according to claim 1, characterized in that: After iterative calculation of all cut-off streams in step (3), when there is an iterative loop inside the model and the calculation requires multi-level iterative interaction inside and outside, the full-flow program solution method is integrated into the iterative solution method at the broken stream, and the residual is used as the objective function to achieve iterative convergence process at different levels.

8. A complex process simulation computing system for process industry, characterized in that: include: Internal iteration module: This module relies on the amount of physical property data, including the thermodynamic data and transfer property data of the material, and is used for the calculation and solution of the basic model. External iteration module: used to perform mathematical modeling and in-depth analysis on the input flowchart to implement any of the steps described in claims 1-7, wherein it can call the calculation and solution results of the internal iteration module as needed.

9. A computer device comprising: Memory; as well as A processor is configured to execute the complex process simulation calculation construction method for process industry as described in claims 1-7 based on the instructions stored in the memory.

10. A computer-readable storage medium storing computer instructions, which, when executed by a processor, can implement the complex process simulation calculation construction method for process industry as described in claims 1-7.

Images